Protein classification method based on SAT and OBDD bucket elimination

Abstract

The invention discloses a protein classification method based on SAT and OBDD bucket elimination, which uses a model of Boolean satisfiability problem (SAT), and utilizes a symbol solving algorithm ofan ordered binary decision graph (OBDD) and a bucket elimination algorithm. The method comprises the following steps: constructing a SAT model by using the constraint relation of the element positions and the cardinality constraint in a candidate mode; using OBDD symbol technology and the included each symbol operation, combining the bucket elimination algorithm and solving the established model,and applying the solving technology to the protein classification, analyzing and extracting the characteristic information in the protein, and classifying the protein effectively. According to the invention, the problem of protein classification is an object, the problem is mined by solving the frequent sequence in pattern mining, and the protein is researched. In the execution process of the algorithm, the search space is effectively reduced, the solving efficiency of the problem is improved, and the algorithm has good practicability.

Description

technical field [0001] The invention relates to the field of data mining and symbol technology, in particular to a protein classification method based on SAT and OBDD bucket elimination. Background technique [0002] Many medically important pathogenic bacteria have an extra "outer" membrane surrounding their cells. The proteins that reside in this membrane (outer membrane proteins OMPs) are prime targets for antibiotic and vaccine drug design because they are located on the surface of bacteria and are therefore the most accessible targets for the development of new drugs. With the development of genome sequencing technology and bioinformatics, biologists can now deduce all the proteins that may be produced in a particular bacterium and try to classify the location of proteins in bacterial cells. However, such protein localization programs are currently the least accurate when predicting OMPs, so better OMP classifiers need to be developed. [0003] Data mining studies hav...

AI Technical Summary

Problems solved by technology

[0005] What the present invention is to solve is the problem that is inevitably restricted by the combination complexity in the process of combining the sequence pattern mining problem with the protein classification problem, and provides a protein classification method based on SAT and OBDD bucket elimination

Images